A single image deblurring approach based on a fractional order dark channel prior

Yu, Xiaoyuan ; Xie, Wei  ; Yu, Jinwei

International Journal of Applied Mathematics and Computer Science, Tome 32 (2022) no. 3, pp. 441-454.

Résumé

The dark channel prior has been successfully applied to solve the blind deblurring problem on different scene images. Since the dark channel of the blurry-noise image is similar to that of the corresponding clear image, the sparsity of the dark channel is less effective for image blind deblurring. Inspired by the fact that a fractional order calculation can inhibit the noise and preserve the texture information of the image, a fractional order dark channel prior is proposed for image deblurring in this paper. It is appropriate for kernel estimation where input images and intermediate images are processed by using a fractional order dark channel prior. Furthermore, the non-convex problem is solved by the half-quadratic splitting method, and some metrics are used for deblurring image quality assessment. Finally, quantitative and qualitative experimental results show that the proposed method achieves state-of-the-art results on synthetic and real blurry images.

Détail

Keywords: blind image deblurring, fractional order dark channel prior, nonconvex problem
Mots-clés : obraz rozmyty, rząd ułamkowy, problem niewypukły

Bibliographie

[1] [1] An, S., Roh, H. and Kang, M. (2020). Long-term residual blending network for blur invariant single image blind deblurring, arXiv: 2007.04543.

[2] [2] Cho, S. and Lee, S. (2009). Fast motion deblurring, ACM SIGGRAPH Asia 28(5): 1–8.

[3] [3] Fergus, R., Singh, B., Hertzmann, A., Roweis, S.T. and Freeman, W.T. (2006). Removing camera shake from a single photograph, ACM Transactions on Graphics 25(3): 787–794.

[4] [4] Fuhai, C., Rongrong, J., Chengpeng, D., Xiaoshuai, S., Chia-Wen, L., Jiayi, J., Baochang, Z., Feiyue, H. and Liujuan, C. (2019). Semantic-aware image deblurring, arXiv: 1910.03853.

[5] [5] Gao, D., Liu, J.,Wu, R., Cheng, D., Fan, X. and Tang, X. (2019). Utilizing relevant RGB-D data to help recognize RGB images in the target domain, International Journal of Applied Mathematics and Computer Science 29(3): 611–621, DOI: 10.2478/amcs-2019-0045.

[6] [6] Gong, D., Yang, J., Liu, L., Zhang, Y., Reid, I., Shen, C., Van Den Hengel, A. and Shi, Q. (2017). From motion blur to motion flow: A deep learning solution for removing heterogeneous motion blur, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, USA, pp. 3806–3815.

[7] [7] He, K., Sun, J. and Tang, X. (2009). Single image haze removal using dark channel prior, 2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami, USA, pp. 1956–1963.

[8] [8] Jia, H. and Pu, Y. (2008). Fractional calculus method for enhancing digital image of bank slip, 2008 Congress on Image and Signal Processing, Sanya, China, Vol. 3, pp. 326–330.

[9] [9] Joshi, N., Zitnick, C.L., Szeliski, R. and Kriegman, D.J. (2009). Image deblurring and denoising using color priors, 2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami, USA, pp. 1550–1557.

[10] [10] Kohler, R., Hirsch, M., Mohler, B., Scholkopf, B. and Harmeling, S. (2012). Recording and playback of camera shake: Benchmarking blind deconvolution with a real-world database, 2012 European Conference on Computer Vision, Florence, Italy, pp. 27–40.

[11] [11] Kotera, J., Smidl, V. and Sroubek, F. (2017). Blind deconvolution with model discrepancies, IEEE Transactions on Image Processing 26(5): 2533–2544.

[12] [12] Kotera, J., Šroubek, F. and Milanfar, P. (2013). Blind deconvolution using alternating maximum a posteriori estimation with heavy-tailed priors, in R. Wilson et al. (Eds), Computer Analysis of Images and Patterns, Springer, Berlin/Heidelberg, pp. 59–66.

[13] [13] Kupyn, O., Budzan, V., Mykhailych, M., Mishkin, D. and Matas, J. (2018). DeblurGAN: Blind motion deblurring using conditional adversarial networks, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, USA, pp. 8183–8192.

[14] [14] Lai,W.-S., Ding, J.-J., Lin, Y.-Y. and Chuang, Y.-Y. (2015). Blur kernel estimation using normalized color-line priors, 2015 IEEE Conference on Computer Vision and Pattern Recognition CVPR), Boston, USA, pp. 64–72.

[15] [15] Lai, W.S., Huang, J.B., Hu, Z., Ahuja, N. and Yang, M.H. (2016). A comparative study for single image blind deblurring, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, USA, pp. 1701–1709.

[16] [16] Levin, A., Weiss, Y., Durand, F. and Freeman, W.T. (2009). Understanding and evaluating blind deconvolution algorithms, 2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami, USA, pp. 1964–1971.

[17] [17] Levin, A., Weiss, Y., Durand, F. and Freeman, W.T. (2011). Efficient marginal likelihood optimization in blind deconvolution, 2011 IEEE Conference on Computer Vision and Pattern Recognition, Colorado Springs, USA, pp. 2657–2664.

[18] [18] Li, B. and Xie, W. (2015). Adaptive fractional differential approach and its application to medical image enhancement, Computers Electrical Engineering 45(C): 324–335.

[19] [19] Li, B. and Xie, W. (2016). Image denoising and enhancement based on adaptive fractional calculus of small probability strategy, Neurocomputing 175(Part A): 704–714.

[20] [20] Li, J. and Lu, W. (2016). Blind image motion deblurring with [...]-regularized priors, Journal of Visual Communication Representation 40(Part A): 14–23.

[21] [21] Li, P., Prieto, L., Mery, D. and Flynn, P.J. (2019). On low-resolution face recognition in the wild: Comparisons and new techniques, IEEE Transactions on Information Forensics and Security 14(8): 2000–2012.

[22] [22] Liu, Y., Wang, J., Cho, S., Finkelstein, A. and Rusinkiewicz, S. (2013). A no-reference metric for evaluating the quality of motion deblurring, ACM Transactions on Graphics 32(6): 175:1–175:12.

[23] [23] Matychyn, I. and Onyshchenko, V. (2021). Time-optimal control of linear fractional systems with variable coefficients, International Journal of Applied Mathematics and Computer Science 31(3): 375–386, DOI: 10.34768/amcs-2021-0025.

[24] [24] Pan, J., Hu, Z., Su, Z. and Yang, M.H. (2014a). Deblurring text images via [...]-regularized intensity and gradient prior, 2014 IEEE Conference on Computer Vision and Pattern Recognition, Columbus, USA, pp. 2901–2908.

[25] [25] Pan, J., Liu, R., Su, Z. and Liu, G. (2014b). Motion blur kernel estimation via salient edges and low rank prior, 2014 IEEE International Conference on Multimedia and Expo (ICME), Chengdu, China, pp. 1–6.

[26] [26] Pan, J., Sun, D., Pfister, H. and Yang, M. (2018). Deblurring images via dark channel prior, IEEE Transactions on Pattern Analysis and Machine Intelligence 40(10): 2315–2328.

[27] [27] Ren, W., Cao, X., Pan, J., Guo, X., Zuo, W. and Yang, M.H. (2016). Image deblurring via enhanced low-rank prior, IEEE Transactions on Image Processing 25(7): 3426–3437.

[28] [28] Shan, Q., Jia, J. and Agarwala, A. (2008). High-quality motion deblurring from a single image, ACM Transactions on Graphics 27(3): 1–10.

[29] [29] Sun, L., Cho, S., Wang, J. and Hays, J. (2013). Edge-based blur kernel estimation using patch priors, IEEE International Conference on Computational Photography (ICCP), Cambridge, USA, pp. 1–8.

[30] [30] Wang, H., Pan, J., Su, Z. and Liang, S. (2018). Blind image deblurring using elastic-net based rank prior, Computer Vision and Image Understanding 168: 157–171.

[31] [31] Wang, Z., Simoncelli, E.P. and Bovik, A.C. (2003). Multiscale structural similarity for image quality assessment, 37th Asilomar Conference on Signals, Systems Computers, Pacific Grove, USA, Vol. 2, pp. 1398–1402.

[32] [32] Chen, X., Yang, Q.W. and Wu, J. (2010). Image deblur in gradient domain, Optical Engineering 49(11): 49–49–7.

[33] [33] Xie, Z. (2016). A primal-dual method with linear mapping for a saddle point problem in image deblurring, Journal of Visual Communication Image Representation 42: 112–120.

[34] [34] Xu, L. and Jia, J. (2010). Two-phase kernel estimation for robust motion deblurring, Proceedings of the 11th European Conference on Computer Vision: ECCV’10, Heraklion, Crete, Greece, Part I, pp. 157–170.

[35] [35] Xu, L., Zheng, S. and Jia, J. (2013). Unnatural L0 sparse representation for natural image deblurring, 2013 IEEE Conference on Computer Vision and Pattern Recognition, Portland, USA, pp. 1107–1114.

[36] [36] Yan, Y., Ren, W., Guo, Y., Wang, R. and Cao, X. (2017). Image deblurring via extreme channels prior, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, USA, pp. 6978–6986.

[37] [37] Yin, M., Gao, J., Tien, D. and Cai, S. (2014). Blind image deblurring via coupled sparse representation, Journal of Visual Communication Image Representation 25(5): 814–821.

[38] [38] Zhang, H., Dai, Y., Li, H. and Koniusz, P. (2019). Deep stacked hierarchical multi-patch network for image deblurring, Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, USA, pp. 5971–5979.

[39] [39] Zhichao, F., Yingbin, Z., Hao, Y., Yu, K., Jing, Y. and Liang, H. (2019). Edge-aware deep image deblurring, arXiv: 1907.02282.

[40] [40] Zhong, L., Cho, S., Metaxas, D., Paris, S. and Wang, J. (2013). Handling noise in single image deblurring using directional filters, 2013 IEEE Conference on Computer Vision and Pattern Recognition, Portland, USA, pp. 612–619.